Lex Fridman PodcastRuss Tedrake: Underactuated Robotics, Control, Dynamics and Touch | Lex Fridman Podcast #114
Summaries, chapters, transcripts, and FAQs on this page are made by uListen, not YouTube.
At a glance
WHAT IT’S REALLY ABOUT
Let Physics Work: Russ Tedrake on Robots, Control, and Touch
- Lex Fridman and MIT roboticist Russ Tedrake explore how control theory, dynamics, and contact shape the behavior of both robots and animals, from passive dynamic walkers to fish surfing vortices and humans running barefoot.
- They dig into underactuated robotics—systems with more degrees of freedom than actuators—and why embracing, rather than fighting, physics leads to more elegant, robust, and efficient robot behavior.
- Tedrake recounts lessons from the DARPA Robotics Challenge, the difficulty of contact-rich manipulation and simulation, and the role of deep learning versus rigorous mathematical thinking in robotics.
- The conversation closes with discussion of soft robotics, tactile sensing, home robots for elder care, societal fears about robots, and advice for young people on thinking deeply and building things that matter.
IDEAS WORTH REMEMBERING
5 ideasLeverage physics instead of fighting it.
Passive dynamic walkers and trout surfing vortices show that if you design mechanics well and stop trying to cancel out the world with motors and control, you can get stable, natural, and efficient motion with minimal actuation.
Underactuated systems demand optimization and careful control design.
Most real systems, including humans, have more degrees of freedom than actuators; optimal control and trajectory optimization let you exploit these dynamics rather than pretend everything is fully actuated and simple.
Contact is both essential and computationally hard.
Making and breaking contact introduces discontinuities and combinatorial explosion in possible contact configurations, which breaks many classic smooth-control assumptions and makes both simulation and planning for manipulation fundamentally challenging.
Rigorous thinking still matters in the age of deep learning.
While deep learning enables fast progress, Tedrake argues that clear mathematical models, analysis, and classical control tools are irreplaceable for safety-critical, complex robotic systems and to avoid over-relying on black-box function approximators.
Robust robotics requires disciplined testing and rare-event search.
The DARPA Robotics Challenge revealed that you cannot simply run a single robot more often; you need systematic testing, unit and integration tests, good simulators, and active methods to expose rare but catastrophic failure modes.
WORDS WORTH SAVING
5 quotesMaybe instead of trying to take the forces imparted to you by the world and replacing them, what we should be doing is letting the world push us around and we go with the flow.
— Russ Tedrake
What if Newton and Galileo had deep learning and just said, ‘Here are the weights of the neural network’? I don’t think we’d be as far as we are today.
— Russ Tedrake
The world gives us underactuated robots, whether we like it or not.
— Russ Tedrake
We’re building robots wrong. Robots are afraid of touching the world all over their body.
— Russ Tedrake
What a luxury to have something that you want to spend all your time on.
— Russ Tedrake
High quality AI-generated summary created from speaker-labeled transcript.
